Early life stress is the most potent predictor of disadvantageous neurobehavioral development in human children. The validation of experimental translational animal models of early life stress will be essential to understanding how early experience affects infant development in positive and negative ways. The development of these strategies will be our best experimental avenue to develop informed intervention and prevention strategies for improving the lives of children that experience early trauma. Some well-characterized experimental animal models of early stress (e.g., maternal deprivation) are difficult to translate to the human experience. Many other manipulations exert their effects by reducing parental investment (e.g., variable foraging demand, repeated maternal separation). These manipulations work well because there are few more potent stressors for the developing infant than poor parental care (e.g., neglect or abuse). A related strategy is to observe the range of care individuals receive and determine its effects on variation in individual development. This is a powerful approach, as it allows us to determine the effects of moderate maltreatment, in addition to more extreme influences, on infant development. A significant drawback to this approach is that shared genes and environment between the mother and infant confound the investigation. Because of the potential for this gene- environment correlation, it i impossible to know, for example, whether the experience of neglect and abuse leads to adverse neurobehavioral traits, or whether shared genes between parent and offspring are associated with both poor parenting and enhanced stress reactivity. Therefore, studies of the effects of early environment must include appropriate cross-fostered (FOSTER) conditions, when one group consists of infants reared by biologically unrelated mothers. If FOSTER dyads exhibit the same relationship between early care and infant neurobehavioral development as biologically related (BIO) dyads, we can focus on environmentally mediated maternal effects. This can only be accomplished in animal models, as experimental adoption studies in humans cannot be executed for ethical reasons, and naturalistic adoption studies are confounded by the fact that adoptive parents may be selected for greater psychological and physical resources than the normal population. Rhesus macaques (Macaca mulatta) are one of the best animal models of neurobehavioral development, as they are closely related to humans, and display comparable genetic, neural, and socio-emotional complexity. Importantly, rhesus macaque mothers, like all mammalian mothers studied to date, vary in the quality of care they give to infants. The only obstacle to the effective use of macaque FOSTER dyads is that some procedures for fostering infants to rhesus macaque mothers may reduce maternal care and enhance infant stress reactivity, making them poor controls for BIO dyads. The PI proposes to determine the effects of natural variation in maternal care quality on rhesus macaque infant development while testing new FOSTER procedures to reduce the differences between FOSTER and BIO dyads. If the relationship between maternal care and infant outcomes does not differ between BIO- and FOSTER- dyads, this will represent a new platform for understanding the genetic, epigenetic, neural and experiential bases of risk and resiliency in the most translatable animal model available.